LBNL-41846

India’s Fertilizer Industry: Productivity and Energy Efficiency

Katja Schumacher and Jayant Sathaye

Energy Analysis Program

Environmental Energy Technologies Division
Ernest Orlando Lawrence Berkeley National Laboratory
University of California
Berkeley, California 94720

July 1999

This work was supported by the Director, Office of Science, Office of Biological and Environmental Research,
Environmental Sciences Division, of the U.S. Department of Energy under Contract No. DE-AC03-76SFOO098.
 ACKNOWLEDGEMENTS

The authors would like to thank Joyashree Roy, Ernst Worrell, Puran
Mongia, and Alan Sanstad for their valuable assistance and comments on
previous drafts of this paper. This work was supported by the
Environmental Science Division, Office of Biological and Environmental
Research (OBER), Office of Energy Research, U.S. Department of
Energy, under Contract No. DE-AC03-76SFOO098.
 DISCLAIMER

This repoti was prepared as an account of work sponsored

byan agency of the United States Government. Neither
the United States Government nor any agency thereof, nor
any of their employees, make any warranty, express or
implied, or assumes any legal liability or responsibility for
the accuracy, completeness, or usefulness of any
information, apparatus, product, or process disclosed, or
represents that its use would not infringe privately owned
rights. Reference herein to any specific commercial
product, process, or service by trade name, trademark,
manufacturer, or otherwise does not necessarily constitute
or imply its endorsement, recommendation, or favoring by
the United States Government or any agwicy thereof. The
views and opinions of authors expressed herein do not
necessarily state or reflect those of the United States
Government or any agency thereof.

.—.——
— ..—.—---- —.—. -. -,.- .. --.-..- r.-.=,-- . . . .. . . . . -- . ..--.-,-. .-— —, .-, .,.
 DISCLAIMER

Portions of this document may be illegible

in electronic image products. images are
produced from the best available original
document.
 Abstract

Historical estimates of productivity growth in India’s fertilizer sector vary from

indicating an improvement to a decline in the sector’s productivity. The variance may be
traced to the time period of study, source of data for analysis, and type of indices and
econometric specifications used for reporting productivity growth. We derive both
growth accounting and econometric estimates of productivity growth for this sector. Our
results show that over the observed period from 1973-74 to 1993-94 productivity
increased by 2.3’-XO p.s. as indicated by the Translog index. Calculations of the Kendrick
and Solow index support this finding. The increase took place during the era of total
control when a retention price system and distribution control was in effect. The retention
price system was coupled with relatively high norms on capacity utilization which
supported productivity increase. With liberalization of the fertilizer sector and reduction
of subsidies productivity declined substantially since the early 1990s, despite an increase
in capacity utilization. Using a translog specification the econometric analysis reveals that
technical progress in India’s fertilizer sector has been biased towards the use of energy,
while it has been capital and labor saving. The analysis of current changes in structure
and energy efficiency in the sector demonstrates the major role policies and fiscal
incentives play in the Indian fertilizer sector. Substantial energy savings and carbon
reduction potential exists. Energy policies can help overcome barriers to the adoption of
energy efficiency measures in giving proper incentives and correcting distorted prices.

.. .
111
 Table of Contents

List of Tables vi

List of Figures vii

1. Introduction 1

2. Fertilizer Industry 2

2.1. The Fertilizer Industry in Context 2

2.2. Fertilizer Process 3

2.3. Fertilizer Production in India 5

2.3.1. Raw Materials 8
2.3.2. Energy Use 9

2.4. Past and Future Demand 10

2.5. Policy 11

3. Statistical and Econometric Analysis 16

3.1. Statistical Analysis 16

3.1.1. Previous Studies 17
3.1.1.1. Partial Productivity 18
3.1.1.2. Total Factor Productivity Growth 21
3.1.2. Own Estimates 21
3.1.2.1. Partial Productivity 21
3.1.2.2. Total Factor Productivity 23
3.1.2.3. Total Productivity 24

3.2. Econometric Analysis 26

3.2.1. Previous Studies 26
3.2.2. Own Estimates 27

3.3. Discussion 28

iv

. .. .. ..—.-. . —---...--- ——- —-..—— . ... ..-..-.—..—--. -. -- . . . -“--- .— . .. .... .

4. Future Development of the Fertilizer Sector 31

4.1. Ongoing Changes in the Fertilizer Industry 31

4.2. Potentials for Energy Efficiency Improvements 32

4.2.1. India versus Best Practice 32
4.2.2. Categories for Energy Efficiency Improvement 33
4.2.3. Barriers to Energy Efficiency Improvement 34

5. Summary and Conclusions 34

References 36

Appendix 38
 List of Tables

Table 2.1
Economic Indicators for the Fertilizer Industry
Table 2.2
Production of Fertilizer by Products
Table 2.3
Fertilizer Production, Capacity and Capacity Utilization
Table 2.4
Fertilizer Production, Capacity and Capacity Utilization by Sector (1991)
Table 2.5
Share of Fertilizer Products in Total Nutrient Production (1990-91)
Table 2.6
Specific Final Energy Consumption of Ammonia Plants: Average and by
Feedstock
Table 2.7 Fertilizer Consumption, Imports and Subsidies
Table 2.8 Demand Projections - Fertilizer
Table 2.9 Overview of Policies Regarding the Fertilizer Industry

Table 3.1 Partial Productivity Growth

Table 3.2 Total Factor Productivity Growth
Table 3.3 Total Productivity Growth
Table 3.4 Decomposition of Growth in Value of Output
Table 3.5 Estimated Parameters for the Translog Cost Function Approach
Table 3.6 Technical Change Bias
Table 3.7 Price Elasticities
Table 3.8 Elasticities of Substitution – Qualitative Overview

Table 4.1 Age, Number, Installed Capacity and Feedstock of Ammonia Plants in India
Table 4.2 Energy Savings Potential – All Ammonia Plants: Broad Estimates
 List of Figures

Figure 2.1 Changes in Physical Energy Intensity of Various Industries

Figure 3.1 Estimates of Total Factor Productivity Growth

Figure 3.2 Estimates of Partial Productivity Growth: Capital
Figure 3.3 Estimates of Partial Productivity Growth: Labor
Figure 3.4 Estimates of Capital-Labor Ratio
Figure 3.5 Index of Partial Productivity
Figure 3.6 Index of Total Factor Productivity
Figure 3.7 Index of Total Productivity

vii
1. Introduction

The fertilizer industry presents one of the most energy intensive sectors within the Indian
economy and is therefore of particular interest in the context of both local and global
environmental discussions. Increases in productivity through the adoption of more
efficient and cleaner technologies in the manufacturing sector will be most effective in
merging economic, environmental, and social development objectives. A historical
examination of productivity growth in India’s industries embedded into a broader
analysis of structural composition and policy changes will help identi& potential future
development strategies that lead towards a more sustainable development path.

Issues of productivity growth and patterns of substitution in the fertilizer sector as well as
in other energy intensive industries in India have been discussed from various
perspectives. Historical estimates vary from indicating an improvement to a decline in the
sector’s productivity. The variation depends mainly on the time period considered, the
source of dat~ the type of indices and econometric specifications used for reporting
productivity growth. Regarding patterns of substitution most analyses focus on interfuel
substitution possibilities in the context of rising energy demand. Not much research has
been conducted on patterns of substitution among the primary and secondary input
factors: Capital, labor, energy and materials. However, analyzing the use and substitution
possibilities of these factors as well as identi~ing the main drivers of productivity growth
among these and other factors is of special importance for understanding technological
and overall development of an industry.

In this paper we contribute to the discussion on productivity growth and the role of
technological change. We introduce the fertilizer industry in more detail taking into
account industry specific aspects such as structural composition, production,
technologies, energy consumption within processes, sector specific policies etc. This
following we derive both statistical and econometric estimates of productivity growth for
the fertilizer sector over time. For the statistical analysis we develop the Kendrick and
Solow indices while for the econometric analysis a translog cost fimction approach using
both cross-state and national time series data is employed. The results are then interpreted
within a broader context of structural and policy changes in the sector as well as other
sector specific aspects.

Future energy use depends on the level of production and the technologies employed.
Furthermore, different economic and policy settings affect structures and efficiencies
within the sector. The final section therefore examines the ongoing changes in the
fertilizer industry structure. It compares best practice technologies to Indian technologies
and identi~ potentials and barriers to the adoption of such efficiency improvements.
2. Fertilizer Industry

2.1 The Fertilizer Industry in Context

Six industries in India have been identified as energy intensive industries: Aluminum,
cement, fertilizer, iron and steel, glass, and paper. Together they account for 16.8’% of
manufacturing value of output (VO) and consume 38.8°/0 of all fhels consumed in the
manufacturing sector (Table 2.1). The fertilizer sector holds a considerable share within
these energy intensive industries. In 1993, it accounted for 23% of value of output within
the six industries and for 3.8’XOin the manufacturing sector.

Table 2.1: Economic Indicators for the Fertilizer Indl

-. -,...-.. ,.,.-,:...-.-,.. :,.>’.
.,,_...).>; -r.-.. -., -::,. , ,, j ,’UIW“:“ ~.Fe@iz&’,,..
>~.,: ,-. ..,~” ..,,..- -;: .:.’.<’;“: ‘ . . .. . . ,,, . .,
.. ‘:+ .’ - :“. . .. =. 7 ‘:..:.-. > . ....
-’‘. ...
-.-,,,-......,.=
: .. ~,,-. . .. . . . . . . . ..
*.,:..., ,.-, ............~.:. ... . . .
,.. . . . ......
... .-..... . . ., . ... . . ..
Groh-cvkue’oi”oui$utl 1
No”minal
1973-1993 0/0 p.s. 17.5 16.4 15.1
1973-1979 0/0 p.s. 22.8 19.2 16.2
1979-1991 0/0 p.s. 16.3 15.8 14.4
1991-1993 0/0 p.s. 8.6 12.2 16.2
Real
1973-1993 0/0 p.s. 10.1
1973-1979 0/0 p.s. 16.4
1979-1991 0/0 p.s. 10.4
1991-1993 0/0 p.s. -10.6
In 1993-94:
VO Share in Aggr. Sector VO/ 3.8?40 16.8% 100%
Manufacturing (nominal) Manuf. VO
Sector Fuel Share in Aggr. Sector Fuel/ 7.4% 38.8?40 100%
Manuf. (nominal) Manuf. Fuel .
Share of Fuel Costs in Sector Fuel/ 13.2% 15.8% 6.8%
Value of Output (nominal) Sector VO
Source: Government of India, ASI (various years).
1calculatedasexponentialannualgrowth.

Production in the fertilizer sector has been increasing over the last 20 years. Over the
study period 1973-1993 real VO increased by an average of 10.10/0 p.s. The fertilizer
industry shows highest growth in the group of energy intensive industries. Major fertilizer
pricing specific policy changes took place in 1977-80 and since 1991. As seen in Table
2.1 growth of real value of output was rising at around 16.4% during the first period
(1973-1 979) and increased slightly lower at 10.4% in the following period of total control
(1979-91) accounting for higher than average growth in both the group of six energy
intensive industries and total manufacturing. After 1991, real value of output decreased
substantially at –1 0.6°/0 until 1993 despite major policy changes towards decontrol and
liberalization that aimed at spurring the sector.
Figure 2.1: Changes in Physical Energy Intensity of Various Industries
(Real Fuel Cost/Real Value of Output -1973-74 values)

0.35
T
. .... .. ... ...
0.30
..-
0.25
...--’ ”””-... ..
..--”-..
........... .. ..- . ..... . ..... ..... ....-.- ..-”---

0.15...?
,.
....●.

0.05..
\ & k

Years

l—paper —Fertilizer —Glass . . . . . . . Cement - +_lmn & Steel . . .. . .. Aluminum —Total Manufactunngl

In 1993-94, the fertilizer sector accounts for 7.4% of total fiels consumed in the
manufacturing sector. Within the group of energy intensive industries, the share of fiels
consumed per unit of output (VO) is second lowest (13 .2°/0) after the iron and steel
industry and thus less than the average of the six energy intensive industries. However,
fiels consumed per unit of output are still about twice the average of total manufacturing
unit fiel costs. Figure 2.1 displays the energy intensity of the fertilizer sector in real
values. The ‘real-value’ indicator reflects the changes in physical energy intensity over
time and gives a comparison to other sectors. Besides iron and steel production, fertilizer
production has been least energy intensive not only in 1993 but almost over the whole
time period. Overall, despite its fluctuating pattern the fertilizer industry shows a
relatively stable slightly decreasing trend in energy intensity.

2.2 Fertilizer Process

Agricultural growth is mainly dependent on advances in fting technologies and

increased use of chemical fertilizers. The fertilizers contain the three basic nutrients for
agriculture: nitrogen (N), phosphorous (P) and potassium (K). Nitrogen is primarily
provided by nitrogenous fertilizers such as urea (46%N) or ammonia fertilizers, e.g.
ammonium sulfate (20.6°/oN). Further shares of nitrogen are contained in complex
fertilizers that combine all three plant nutrients (NPK). Phosphate comes in the form of
straight phosphatic fertilizers such as single super phosphate (16°/0P20~) or as part of a
complex fertilizer. Potassic fertilizer is available as straight potassic fertilizer, such as

3
muriate of potash (60°AK20) or sulfate of potash (50?40K20) or as a complex NPK
fertilizer component.

The effectiveness of fertilizers can only be assured if applied in optimal combination

specific to the local soil and climatic conditions. Nitrogen presents the most essential
nutrient for plant growth holding the biggest share in the optimal mix. The basic raw
material for the production of nitrogenous fertilizers is ammonia, for straight phosphatic
fertilizers it is phosphate and for potassic fertilizers potash. Out of the three fertilizer
types, production of ammonia is most energy and resources intensive. Its production
process is presented in more detail here. The description draws on Phylipsen, Blok and
Worrell (1998).

The most important step in producing ammonia (NH~) is the production of hydrogen,
which is followed by the reaction between hydrogen and nitrogen. A number of processes
are available to produce hydrogen, differing primarily in type of feedstock used.

The hydrogen production route predominantly used world wide is steam reforming of
natural gas. In this process natural gas (CHO) is mixed with water (steam) and air to
produce hydrogen (H2), carbon monoxide (CO) and carbon dioxide (C02). Waste heat is
used for preheating and steam production, and part of the methane is burnt to generate the
energy required to drive the reaction. CO is further converted to C02 and Hz using the
water gas shift reaction. After CO and COZ is removed from the gas mixture ammonia
(NH~) is obtained by synthesis reaction.

Another route to produce ammonia is through partial oxidation. This process requires
more energy (up to 40-50°/0 more) and is more expensive than steam reforming. The / I
advantage of partial oxidation is a high feedstock flexibility: it can be used for any
gaseous, liquid or solid hydrocarbon. In practice partial oxidation can be economically
viable if used for conversion of relatively cheap raw materials like oil residues or coal. In
this process air is distilled to produce oxygen for the oxidation step. A mixture containing
among others H2, CO, COZ and CHd is formed. After desulfi.u-ization CO is converted to
C02 and H20. COZ is removed, and the gas mixture is washed with liquid nitrogen
(obtained from the distillation of air). The nitrogen removes CO from the gas mixture and
simultaneously provides the nitrogen required for the ammonia synthesis reaction.

A variety of nitrogenous fertilizers can be produced on the base of ammonia. Ammonia

can be used in a reaction with carbon dioxide to produce urea. Ammonia nitrate can be
produced through the combination of ammonia and nitric acid adding fiu-ther energy in
form of steam and electricity. Other fertilizer types produced on the base of ammonia
include calcium ammonium nitrate (ammonium nitrate mixed with ground dolomite) and
NP/NPK compound fertilizers. For the most part fhrther energy is required to induce the
necessary chemical processes.
Phosphatic fertilizers are produced on the basis of phosphoric and suhr.ic acids.
Phosphoric acid is produced by leaching of phosphate rock with sulfuric acid. Sulfic
acid very often remains as a waste product of the chemical industry. (Worrell et al., 1994)

Potash fertilizers are produced from sylvinite salt. Sylvinite is diluted in a circulation
fluid in the flotation process. The potash fertilizer is separated by skimming the solution.
(Worrell et al., 1994)

2.3 Fertilizer Production in India

India is the fourth largest producer as well as consumer of fertilizer in the world. With
population growing at a fmt rate, food production was given highest priority in India
since the 1960s (New Agricultural Strategy). Although India’s soil is varied and rich, it is
naturally deficient in major plant nutrients (nitrogen, phosphate and potassium). Growth
in chemical fertilizer production and consumption therefore presents the single largest
contributor to agricultural progress, its technological transformation and
commercialization.

Table 2.2: Production of Fertilizer by Products (’000 tonnes)

Year . -’::.,, . .,: ‘-: : -. ~98!7--:.198;. 1989- ‘ ,.1990-~~~~
.-. ..;..:1991- ~~
~1992- .-1993-: ;1994-.
.,. 88..
,-. 89...: ~~90: ; “~“, 913-’”:;..92-:+? ~~$3+-~’94 ‘. <% ‘~-
Nitrogenous Fertilizers 5,467 6,712 6,747 6,993 7,302 7,431 7,231 7,944
S~aight Nitrogenous 4,764 5,729 5,991 6,148 6,156 6,321 6,376 6,801
Fertilize
AmmoniumSu~ate 539 610 586 558 553 563 620 585
(20.6%N)
Urea(46%N) 9,835 11,867 12,486 12,836 12,832 13,126 13,150 14,137
CalciumAmmonium 421 480 425 436 446 546 666 572
Nitrate (25%N)
AmmoniumChloride 87 97 80 79 112 122 131 137
(25%N)
Nitrogenous Fertilizer 703 984 757 845 1,145 1,110 855 1,144
throwzh NP/NPK
Phosphatic Fertilizers 1,666 2,253 1,795 2,051 2,562 2,321 1,874 2,557
Straight Phosphatic 398 471 502 584 478 373 361 484
Fertilize
SingleSuper 397 471 502 584 478 373 361 484
Phosphate(16%PIO~
Phosphatic Fertilizer 1,268 1,781 1,293 1,467 2,084 1,948 1,513 2,073
through NP/NPK
Source: Center for Monitorin$ le Indian Economy (1996).

Fertilizer production in India has been growing at an accelerating rate, fi-om very low
levels after independence (0.04 million tonnes] in 1951) and still low levels in the early
1970s to a total production of 11.36 million tonnes in 1995 (Table 2.3). Currently, India
produces various kinds of both nitrogenous and phosphatic fertilizers domestically (see

[ metric tonnes, sometimes abbreviated as ~ or million tonnes as Mt in the following.

. . ....——. . .-.—— —..-——.....-—. .——— . .. . . . .. . . —.—— .-. . .-— — ...c - .. .——-—-.—.—. -----
Table 2.2). These include straight nitrogenous fertilizers (urea and ammonium), straight
phosphatic fertilizers (single super phosphate) and complex fertilizers (like di-amrnonium
phosphate (DAP)). Potassic fertilizers are not manufactured domestically due to lack of
indigenous reserves of potash, the main input. The capacity of nitrogenous fertilizer has
almost doubled with the commissioning of large sized gas based fertilizer plants in the
1980s. Capacity utilization has increased considerably from around 60% in 1979-80 to
over 90’XOin the 1990s.

External factors such as weather and monsoon conditions as well as policy changes
regarding fertilizer production, use and agricultural output enhancement exert significant
influence on capacity utilization in the industry. For instance, following the decontrol of
price and distribution of potassic and phosphatic fertilizers in 1992 capacity utilization
for phosphatic plants fell to 66.3% in 1993-94 from a high at 92.4% in 1991-92.
However, immediately thereafter capacity utilization improved remarkably again to
90.5% in 1994-95.

Table 2.3: Fertilizer Production. . Car. lcity and Capacity Utilization (million tonnes of nutrients)
Year Production Capacity ,. . . ‘~apaci~ Utilization
N P205 ‘Total N P20, Total N P20~ Total
1973-74 1.05 0.32 1.37
1979-80 2.22 0.76 2.98 3.90 .— 1.28 5.18 56.9% 59.4% ——. 57.5?40
—
1985-86 4.32 – 1.43 5.75 5.92 1.72 7.64 if.(itiO
— 83.1% 75.3’%
1986-87 5.41 1.44 6.85 6.76 2.32 9.08 80.0% 62.1% 75.4%
1987-88 5.47 1.67 7.14 7.08 2.47 9.55 77.3% 67.6% 74.8’%0
1988-89 6.71 2.25 8.96 8.16 2.67 10.83 82.2% 84.3% 82.7%
1989-90 6.74 1.79 8.53 8.15 2.72 10.87 82.7’% –—-–.
85.8% 65.8% - 78.5%
1990-91 — 6.99 2.05 9.04 8.15 2.75 10.90 74.5~o 82.9~o”
1991-92 7.30 2.56 9.86 8.21 2.77 10.98 88.9% 92.4% 89.8’%
1992-93 7.43 2.32 9.75 8.51 2.81 11.32 87.3% 82.6% 86.1%
1993-94 7.23 1.87 9.10 8.51 2.82 11.33 85.0% 66.3% 80.3%
1994-95 7.94 2.56 10.50 8.84 2.83 11.67 89.8% 90.5?40 90.OVO
1995-96 8.77 2.59 11.36 9.00 2.92 11.92 97.4% 88.7% 95.3%
Source: Corn ;d fromTrivedi[1998) and Mittal (1994).
, I
In 1991, 47% (49% in 1995) of installed capacity was held by public sector units. The
private sector accounted for 36% (35Y0 in 1995) and the co-operative sector for only 17%
(16% in 1995) .(Table 2.4; CMIE, 1996) Production shares are distributed slightly
differently, due to sector specific capacity utilization and efficiencies. Public units have
lower capacity utilization (Table 2.4). While most of the nitrogenous fertilizer production
capacity can be found in the public sector, phosphatic fertilizer capacity is mainly
installed in the private sector.

6
Table 2.4: Fertilizer Production, Capacity and Capacity
Utilization bv. Sector (1991) . . (million tonnes of nutrients)
,,
.. .<
., .: ..:., .. ‘; ..,., ““ ‘Productioti,’. -~~’
.,,– - ..- -+ ,.-.. + :,’”
.... -,.
....,“–
~,,:- ..,::
.. ”:,,.....:“‘Capacity-,---~:~,;~apaciiyiy
. ....:.-.:
,-.,. ..-,.
, ;.-., . ..., ‘~~~
,., :, . ..:,.:..-.>..,-.. . ..:. ,.. . .. . . ,,, .: Utili&ition--,
Nitrogen (* ‘ “’”7.30 “’8.21 89.0’%
Share in N:
Public Sector 41.4% 52.9% 69.6’%
Private Sector 35.170 28.5% 109.4%
Coop. Sector 23.7% 18.6’%0 I 13.170
Phosphate (P) 2.56 2.77 92.4%
Share in P:
Public Sector 28.5’% 28.9% 91.3%
Private Sector 57.8% 59.9’% 89.2%
Coop. Sector 13.7’70 11.2% 112.9%
Total 9.86 10.98 89.8’XO
Share in Total:
Public Sector 38.0% 46.8% 73.0’%0
Private Sector 41.0’70 36.4’XO 101.0%
Coop. Sector 21.lVO 16.8% 113.0%
Source: Mittal (1994).

For nitrogenous
— fertilizer capacity the share of the public sector has been declining over
time. In as early as 1960-61 ‘the public sector accounted for 87% of nitrogenous fertilizer
capacity. The private sector held a share of 13°/0 and the co-operative sector did not exist
yet. With the introduction of co-operative units and policy changes towards greater
investment in the private sector (induced by a system of retention prices) in 1977, the
share of the public sector started to decline and that of the private and co-operative sector
to improve. Regarding phosphatic fertilizer production, throughout the years, the private
sector has always enjoyed the highest share of up to 64%. (Mittal, 1994)

Today, the cooperative sector assumes a major role within the fertilizer industry not only
in fertilizer distribution but also in the provision of other general services to farmers such
as credit programs, capital management, training schedules, etc. Around 97°/0 of the
villages and 45°/0 of the rural population are participating in cooperative systems.

As can be seen in Table 2.5 about 88% of nitrogen nutrient come from straight
nitrogenous fertilizer products, mainly from urea. The remaining share is covered through
the use of complex fertilizer such as DAP and others. Phosphates are drawn from straight
phosphatic fertilizers to only 28%. The main share is supplied by complex NP/NPK
fertilizers. The role of organic manure (green fertilizer) as a substitute ardor complement
to chemical fertilizers has not yet found much consideration. Cow dung is widely used as
fbel in rural areas, while fi.uther usefil excrements from cattle as well as compost from
urban waste, from forest litter and other waste materials are not utilized at all.

7
 Table 2.5: Share of Fertilizer Products in Total Nutrient Production (1990-91)
;~etii?er?fojduct j.:; .-{<~ ,;;:~j.’;’{’.;~j~fjh-~of Nitrogen-i.?: SI@e ot2Plio6p@ate:: .....,
“,. ... .. :“,~’-,”.. .L:,:’. ,..~.:,, ,,.,,::,.,. .,.. .,..,-~
. .:’.
,. ,.,- ., .. ., :-----.’-;.:.’.----.’.. - . :::.@$:;:
.:..... .-”..,.,. . :...,‘--’:.$:;.:.
:~,-:.,’ .. . .::>., -;,, ::,:
“.: ->e’..:.?Y;’,@*g;)
1. Straight nitrogenous 87.8’%
Urea 84.4%
Others 3.4%
2. Straight phosphatic 28.5’%.
Single superphosphate 28.5%
Others
3. NP/NPKs (complex) 12.2% 71.5%
DAP 4.9% 42.7%
Others 7.3% 28.8%
Total (1+2+3) 100% 100%
Source: Mittal (1994).

Although the Indian fertilizer sector progressed considerably over the past, there are
various problems associated with the sector. These problems mainly relate to investment
into capacity upgradation and expansion, to profitability of operation, and to availability,
storage and transportation of raw materials and finished products. Investment projects
have been very slow in the past. The delay in setting up fertilizer plants after the issue of
the letter of intent can be up to 8-9 years. This increases the cost of projects considerably
contributing to the continuous problem of capital scarcity. Major investment would need
to be brought in from abroad to meet the capital requirements. The government of India
granted concessions to attract foreign capital inflow for the fertilizer industry, such as
majority equity participation, distribution rights etc. However, due to uncertainties
surrounding the availability of raw materials in India as well as the high profitability of
exports to India, the reaction to these concessions was very low. Capital remained scarce.

Particularly following the oil price shock, costs of inputs to fertilizer production, .
especially for energy as the single main input, were exceptionally high. Moreover,
although the government promoted the setting up of fertilizer plants in dispersed areas to
best possible serve local needs, constraints in transportation and storage of fertilizer still
hit the industry. Availability of storage and transportation capacity was constrained
among other reasons due to shortage in wagons for rail transportation, high demand for
transportation by other sectors, storage capacity, quality and management of stored
products.

2.3.1 Raw Materials

Domestic raw materials are available only for nitrogenous fertilizers. For the production
of urea and other ammonia based fertilizers methane presents the major input which is
gained from natural gas/associated gas, naphtha, fiel oil, low sulfur heavy stock (LSHS)
and coal. In the more recent past, production has more and more switched over to the use
of natural gas, associated gas and naphtha as feedstock. Out of these, gas is most
hydrogen rich and easiest to process due to its light weight and fair abundance within the
country. However, demand for gas is quite competitive since it serves as a major input to

8
 electricity generation and provides the preferred fuel input to many other industrial
processes.

For production of phosphatic fertilizer most raw materials have to be imported. India has
no source of elemental sulfur, phosphoric acid and rock phosphate. Yet, some low grade
rock phosphate is domestically mined and made available to rather small scale single
super phosphate fertilizer producers. SuIfhr is produced as a by-product by some of the
petroleum and steel industries.

2.3.2 Energy Use

Fertilizer production is one of the most energy intensive processes in the Indian industry.
Energy is consumed in the form of natural gas, associated gas, naphth~ fiel oil, low
sulfhr heavy stock and coal. The choice of the feedstock is dependent on the availability
of feedstock and the plant location. It is generally assigned to the plants by the
government.

Production of ammonia has greatest impact on energy use in fertilizer production. It

accounts for 80°/0 of the energy consumption for nitrogenous fertilizer. The feedstock mix
used for ammonia production has changed over the past. Since new capacity in the form
of gas based fertilizer plants was added in the 1980s the share of gas has increased
substantially. In 1992-93, the shares of feedstocks in ammonia production were: 54.2°/0
natural gas, 26.1°A naphth~ 18.2% fuel oil, and 1.5°A coal (TEN, 1996) while, in 1981-
82, it was: 52% naphth~ 19% fuel oil, 19% coke oven gas and 10% coal (Kalr% 1989).
The shift towards the increased use of naturalkissociated gas and naphtha is beneficial in
that these feedstocks are more efficient and less polluting than heavy fhels like fiel oil
and coal. Furthermore, capacity utilization in gas based plants is generally higher than in
other plants. Therefore, gas and naphtha present the preferred feedstocks for nitrogenous
fertilizer production.

Energy intensity in India’s fertilizer plants has decreased over time. This decrease is due
to advances in process technology and catalysts, better stream sizes of urea plants and
increased capacity utilization. Capacity utilization is important as losses and waste heat
are of about the same magnitude no matter how much is actually produced in a plant at a
specific point of time. The evolution of specific energy consumption on average and by
feedstock is shown in Table 2.6. Since ammonia production holds the highest share of
energy consumption, the numbers given here are for energy intensity in ammonia plants.
Actual energy consumption in a plant depends on the age of the technology and the scale
of the plant. For example, a typical ammonia plant established in 1970s would be a 600
tpd gas based process with an efficiency of 9.8 to 10.2 Gcal/Mt. A plant established in
early 1990s would consume only 8.0 to 8.5 GcaUMt. (Trivedi, 1998)

The production of phosphatic fertilizer requires much less energy than nitrogenous
fertilizer. Depending on the fertilizer product, in 1993-94, energy consumption varied
from negative input for sulfuric acid to around 1.64GJ/tonne of fertilizer for phosphoric

-.-.
Tmn ~ ..m-,.,..lm -.. . .Tr.. . - . , .?---- -, ...>...-,. . . . . ,. . . . . . . . . .
acid (TERI, 1996). For sulfuric acid the energy input is negative since more steam (in
energy equivalents) is generated in waste heat boilers than is needed as an input.

Table 2.6: Specific Final Energy Consumption of Ammonia Plants: Average and by
Feedstock (Gcal/Mt of ammonia)
.Fee@ock ”,, .:-,,y” >~~...3,979,-’;1983+
;.- $- ‘l!&6-; ‘:~~88+.
..- ..,:,:? :j.19~1-:,j1992<’ :~.993-;,.-l99~. 1995-
~6 .
.,’. . ,“, ..80,, . .,’~4;;,’,,. 87 ‘89’.;:..
, 92 :’93 ‘ ., .94’ 95
Gas “’ na na na 10.0 9.6 9.5 9.36 9.26 na
Naphtha na na na 12.2 11.7 11.7 11.5 11.3 na
Fuel Oil na na na 14.3 13.5 13.9 14.2a 14.3’ na
Coal na na na 33.8 39.7 42.7 39.7 na na
Average 14.8 14.0 13.4 11.9 11.6 11.4 11.5 11.0 10.9
Source: TERI (1996); Trivedi (1998).
‘includingLSHS;na – not applicable.

2.4 Past and Future Demand

Fertilizer consumption depends on various factors. These include agricultural related

factors such as geographical aspects, calamities, rainfall and irrigation patterns, soil
quality, farming methods, availability of technology and information, varieties and
qualities of seeds as well as access to capital and other inputs. Additionally, fertilizer
consumption depends on more macro oriented factors such as market forces and policies
regarding demand and supply.

The introduction of high yielding varieties of seeds and the greater awareness of the
benefits of fertilizers - spread out through government initiated extension networks that
started in the 1960s - significantly spurred the production and consumption of fertilizers.
As shown in Table 2.7 fertilizer consumption more than doubled between 1980 and 1990.
Imports during the same time period did not increase. Rising consumption could entirely
be met by increases in production. The increase in consumption and production in the
1980s was made possible to a large extent by tremendous subsidies provided by the
government. Between 1980 and 1990 subsidies to fertilizer production increased more
than eightfold. From 1990 on, consumption as well .as imports of fertilizer increased
again accompanied by a fiu-ther increase of subsidies. However, the consumption of
phosphatic fertilizers which can not be produced domestically remained rather stagnant. ,

The optimal mix of fertilizer components depends on the variety of seeds to be grown and
the soil quality specific to the region. To assure more efficient use of fertilizer the
government has promoted the setting up of soil testing laboratories throughout the
country. However, more recently in the progress of liberalization, industry policy and
subsidy schemes moved towards supporting nitrogenous fertilizers relatively more than
phosphatic and potassic fertilizers. (Prasad et al., 1994; Subbiah, 1994) This has led to a
shift in the consumption of nutrients away from a generalized ideal fertilizer balance of
4:2:1 NPK to a ratio of 5.9:2.4:1 NPK in 1991-92 and fhrther to 8.5:2.5:1 NPK in 1995-
96 reducing the economic and ecological productivity of fertilizer substantially.

10
 Demand projections for fertilizer consumption up to the year 2001-02 are presented in
Table 2.8. Due to continued increase in population and food requirements, demand for
fertilizer is expected to fh.rther grow at a optimistically stable mix of nutrients (6.8:2. 1:1
NPK). This will help to substantiate agricultural growth.

Table 2.7: Fertilizer Consumption, Imports and Subsidies

,,Ye~’.$ .~Consqptioq; ~,;?..Y.’.
‘Imports=..-,~*
.,. —-.:’
--- .- ‘.-’. , ,_L:
‘ ,.,,, y,. ~ .,. ;,..,<:
:,..- :Subsidies.F.
--, >... . ... .. .::.
[ ‘.. . ;:,: ......, .... , :,(mi~ionto~es~;,:.. (rnfiliorito~es) ;:.;.:(~:.mil~oti).;.
,; .,,.
1970-71 ‘ 2.26 0.63 na
1980-81 5.52 2.76 5,050
1990-91 12.55 2.76 (1.33) 43,890
1991-92 12.73 2.77 (1.24) 48,000
1992-93 12.16 2.98 (0.86) 57,960
1993-94 12.37 3.17 (1.28) 43,990
1994-95 13.83 3.12 52,410
1995-96 na na 62,350
1996-97 na na 60,930
“--- -- “---- c- .,- ...- . .,-. r- ~,... r - –..-...-. /,,mfi/\
~ Source: ~en~er ror lvlonnormg me mamn Economy (I YYOJ
TWmbersin parenthesesgive import(consumption)of potassicfertilize na – not applicable.

Table 2.8: Demand Projections – Fertilizer (million tonnes of nutrients)

Year “., . . . Nitrogen (TJ) ~~.PhospIiate.(P20J 2.ZPotassium (IW)- ~ - -.’ Total- .“
1997-1998 11.0 3.46 1.55 16.01
1998-1999 11.4 3.56 1.65 16.60
1999-2000 11.7 3.64 1.72 17.10
2000-2001 12.1 3.71 1.77 17.57
2001-2002 12.4 3.77 1.82 18.03
AAGrowtha
(1997-2001) 3.0% 2.2’MO 4.1% 3.O’MO
Source: Trivedi (1998).
‘AAGrowth– averageannualgrowthrate calculatedas compoundgrowth. -

2.5 Policy

The Indian fertilizer sector has been under strict government control for most of the
period since independence. A price and distribution control system was considered to be
necessary not only to ensure fair prices and equal distribution all over the country but also
to provide incentives for more intensive use of fertilizers. A control system of licensing
and approval of collaboration aimed at standardizing technology and capacity of plants.
The goal of government intervention was to improve agricultural productivity and thus
the basic supply of food

Until 1970, only straight nitrogenous fertilizers were under price control and no general
distribution control was practiced. In October 1970, Indian fertilizer policy was
completely reviewed and controls on prices and distribution of fertilizer reintroduced,
reinforced or revised. A control on fertilizer distribution was reintroduced in 1973
(Fertilizer Movement Control Order) due to shortages of supply in various areas. The
movement of fertilizer was subject to the ‘Essential Commodity Act’ (ECA). Percentages

11
and types of fertilizer under control of ECA varied over time depending on the actual
demand and supply requirements.

The price control that had earlier been introduced for four nitrogenous fertilizers was
supposed to cover the costs of individual fertilizer plants and the capacity of the cultivator
to bear the price burden. Therefore, a common controlled price for each type of fertilizer
whether indigenously produced or imported was adopted. However, the retention prices
to different plants in public and private sectors varied, depending on their cost
differential. Prices were paid to a “Common Fertilizer Pool”. Out of this pool different
prices were paid to individual manufacturers and for imports. (Mittal, 1994)

Under the control system units were obliged to use specific types of feedstock assigned
by the government on the basis of the location of the plant and the availability of
feedstock. The choice of feedstock presents an important factor for the capital and
operating costs of a plant. Total cost of production for nitrogenous fertilizers is lowest for
plants based on natural gas or naphtha, substantially higher for fuel oil based plants and
highest for coal based plants. Since feedstocks were assigned, plants produced at very
different costs of production for reasons not internal to the plants or to the efficiency of
their operations. Therefore individual retention prices differed substantially.

Although retention prices were individually based on normative costs and included a
specific return on investment plants, mostly public sector plants, suffered from losses due
to higher actual costs. High costs were due to lower than anticipated capacity utilization
and increases in the prices of raw materials. The oil price shock in 1973 led to
tremendous price increases for imported fertilizers. Additionally, import prices for oil and ,
naphtha rose thus leading to higher indigenous costs of fertilizer production.
Consequently, in 1974, the government of India introduced the Fertilizer Pool
Equalization Charge (FPEC). Indigenous manufacturers were required to contribute a
specific amount, Rs. 610 per tonne of urea, into this pool which was used to subsidize the
cost of imported fertilizers.

Nevertheless, the fertilizer industry suffered following the oil price shock. Cost increases
in the production and import of fertilizer led to a slowing of investment into new
capacity. In order to ensure sufficient use of fertilizer at reasonable prices and to stimulate
investment, in January 1976, the government setup a committee (Marathe Committee) to
revise and improve the pricing system. The major goals of the study conducted by the
committee included the establishment of norms for determining the production costs for
individual units, the rationalization of prices of feedstocks and other inputs, the revision
of ex-factory realization from time to time due to any change in cost of inputs, and the
development of a pricing policy of imported fertilizers.

The committee recommended a 12 percent post tax return on net worth (regardless of the
location, age, technology and cost of production) and fixed a retention price for each unit
according to the costs of production. The computation of cost was based on the
assumption of 80% capacity utilization (for ammonia-urea plants) coupled with

12
consumption norms of raw materials, energy etc. These numbers were fixed with the
view of enhancing capacity utilization, promoting efficiency of existing plants as well as
stimulating investment into new capacity. The system was to be applied to all
manufacturers whether in the public, private or cooperative sector.

In 1977, nitrogenous fertilizers were put under the new scheme of retention prices
involving subsidization as recommended by the committee. According to the nature of
the scheme, subsidies were higher for high cost units and lower for low cost units.
Controlled selling prices of fertilizers were not disturbed. The retention price scheme
encouraged fertilizer consumption through low subsidized fhrm gate fertilizer prices and
on the other hand encouraged production of fertilizer through ensuring adequate returns
to producers. The system was originally thought to be self financing. However, low
consumer prices, rising input costs, increased output and high capital costs of production
resulted in an escalating need for additional governmental subsidies to sustain the system.

The system of freight equalization was applied to the fertilizer industry to ensure supply
of fertilizer at uniform prices throughout the whole country. Further subsidies to the
fertilizer industry were given to support fertilizer use in backward, hilly, inaccessible and
tribal areas and by small and marginal farmers in dry areas. In view of these subsidies
retail prices could be reduced between 1974 and 1979 and remained stable thereafter for
the decade of the 1980s. Due to mounting subsidies and rising profits of fertilizer
companies the government tightened the norms on capacity utilization and depreciation to
reduce retention prices and thus subsidies. In January 1989 with retrospect to April 1988,
capacity utilization norms were increased to 90°/0 for gas-based urea plants and 85°/0 for
naphtha and fiel oil based plants. Depreciation was based on 20 year lifetime instead of
10. Capacity utilization norms were set slightly lower from the 11* year of plant lifetime
onwards: 85°/0 for gas-based urea plants and 80°/0 for naphtha and fuel oil based plants.
Coal based plants were subject to a norm of 60% capacity utilization for the first 10 years
and 55°A thereafter, phosphoric acid plants’ norms were 75’XOand 70°A respectively.

In addition, substantially increasing stocks of fertilizer in the 1980s gave way to consider
abandoning the distribution control system under ECA and introducing a system of free
distribution under the retention price subsidy and a normative transport allowance. The
goal of exempting large quantities of fertilizer from allocation control was to increase
competitiveness among producers and bring down prices as well as to reduce subsidies on
transportation. The government partially eased distribution of fertilizers in 1987 and
1988. Specific quotas not covered by ECA were allowed to be freely distributed.
However, for these quantities the system of freight equalization would not apply with the
effect that extra freight cost involved above the normal freight under ECA allocation
would not be reimbursed.

This following and unlike most other sectors, the industry started resisting against
governmental steps towards liberalization of the sector. Abandoning or reducing
allocation controls of fertilizer, they feared, would diminish fertilizer use, impede
agricultural progress and suspend growth of the fertilizer industry. Furthermore, fertilizer

13
manufacturers would have to transport their products to wherever demand exists
including remote and inaccessible areas. This would place further burden on the overall
strained transport system and generate additional production costs not compensated by
the government.

In order to further reduce subsidies the government decided in July 1991 to increase
fertilizer prices for fmmers by 40Y0. However, as a result of immense pressure by various
lobbies, in August 1991, the government reduced the price increase to 30% for big
farmers and withdrew it for small and marginal farmers. Since the system of dual pricing
included lower prices for small and marginal farmers and higher prices for big farmers
many large fmers started splitting up their holdings into smaller entities to quali~ for
the lower prices.

Following this and the devaluation of the rupee in 1991 subsidies to the industry
mounted. Of all subsidies distributed by the government the fertilizer industry received
more than two-fifth in 1991-92. Moreover, the industry enjoyed much higher subsidies
because of additionally benefits from indirect subsidies through the system of differential
pricing.

As a consequence of increasing budget deficits a system of partial decontrol was

introduced effective August 1992. Prices, distribution and movement (freight
equalization) was completely decontrolled for phosphatic and potassic fertilizers.
However, to fiuther ensure and promote the dispersed use of nitrogenous fertilizers, urea
continued to remain under price control. Its price was reduced by 10O/O.Low analysis
nitrogenous fertilizers which had been decontrolled in July 1991 were brought back under
control. Additionally, the government decided to adjust foodgrain procurement prices to
account for the changes in fertilizer prices.

Immediately following the decontrol of phosphatic and potassic fertilizers prices went up
sharply (70-1 00°/0) and temporary shortages occurred. As a neutralization measure the
government announced (i) the exemption of phosphoric acid from custom duty when
imported for use in fertilizer production, (ii) the exemption of 15’XObasic duty on import
of fertilizer projects under a specific import scheme, (iii) the permission to import raw
materials at ofilcial exchange rates etc. However, these measures did not result in ,

sufficient reduction of fertilizer (particularly DAP) prices. The government had to

provide fhrther subsidies. It became apparent that higher fertilizer prices result in reduced
input use and lower production. Any price adjustment would therefore need to include a
mechanism that compensates farmers for their increased production costs by ensuring
higher product prices. With the decontrol of low analysis nitrogenous fertilizers in June
1994 today all fertilizers except urea have become free of controls. (Datt et al. 1998;
Mittal, 1994; Basu et al., 1994; Singh et al., 1994 and Ahluwalia 1985, 1991)

14
Table 2.9: Overview of Policies Regarding the Fertilizer Industry
Period . ‘- Policy .;’>; :::.” :“”’, “ :“~’!specifics: ;:,” ;;::,?::-:,::-.::’::; ;‘: :,>:-: “;=:’; ::: _: - ‘:”’’’;:,.:;.’::: ‘,
1957 FertilizerControlOrder Price and Distribution Control
Before 1970 PriceControl Control of straight nitrogenous fertilizers including urea,
ammonium sulfate and CAN (1966), no or irregular price controls of
other fertilizers
Distribution Control Gradual decrease of distribution control, no area-wise price and
distribution control (ceiling price only) since 1968
Oct. 1970 Review of Fertilizer Policy
Around 1970 Common Fertilizer Pool Pool used to account for individual cost structure of plants (different
retention prices)
1973 Re-Introduction of FertiIiser Movement Control Order, Essential Commodity Act
Distribution Control (ECA): Percentage fertilizer under ECA varied from time to time
1974 Fertilizer Pool Equalization Additional charge to pool used to subsidize high costs of imported
Charge (FPEC) fertilizer following oil price shock
January 1976 High Power Fertilizer Develop suitable price system: Ensuring higher fertilizer use,
Prices Committee financial health and growth of the industry
1977 Review of Price Control Review of price control for nitrogenous fertilizers based on the
recommendations of the committee
1977 Retention Price System. Introduction of retention price system for nitrogenous fertilizers
Feb. 1979 Price Control Retention price system (RPS) for phosphatic fertilizers such as DAP
and other complex fertilizers (PJP and NPK)
June 1980 Price Decontrol Price Decontrol of low analysis nitrogenous fertilizers (AS, CAN,
ACI)
May 1982 Price Control Retention price system for Single Superphosphate (SSP)
September 1984 Price Control Price control again for low analysis nitrogenous fertilizers
1987-88 Free Allocation of Specific Free movement of quantities of single super phosphate (SSP) and
Quantities of Fertilizer for complex fertilizers not covered by the ECA - No reimbursement of
Rabi (1987-88) and Kharif extra freight cost invoIved over and above the normaI freight under
(1988) ECA allocation.
April 1988 Revision of Price Control Reduction of retention prices and subsidies for nitrogenous fertilizer
July 1991 Price Decontrol Price decontrol of low analysis nitrogenous fertilizers
August 1991 Dual Pricing Policy 30% price increases of fertilizer for big farmers, no price increase
for small and marginal fmers
August 1992 Partial Decontrol Decontrol of prices, distribution and movement of phosphatic and
potassic fertilizer, recontrol of low analysis nitrogenous fertilizers,
10’%o price reduction for urea fertilizer
1992 until Subsidy Subsidies on DAP, NP, NPK fertilizers
March 1993
September 1992 Import Liberalization Import of raw material for manufacture of DAP and other complex
phosphatic fertilizers (not SSP) allowed at lower ofilcial exchange
rate (under dual exchange system) with no customs duty
1993 Subsidy Continuation of subsidies on domestic phosphatic fertilizers,
introduction of subsidies on SSP
June 1994 Decontrol Decontrol of low analysis nitrogenous fertilizers again
Price Increase Urea prices raised by 20%
Source: Datt et al :1998), Mittal (1994), Basu et al. (1994), Singh et al. (1994) and Ahluwalia (1985, 1991).

15

..
3. Statistical and Econometric Estimates

3.1 Statistical Analysis

A variety of studies on productivity growth and technological change in Indian industries

has been carried out so far. Originally these studies were driven by an interest in
understanding the capital vanishing phenomena in the Indian industry between 1950 and
1980. During that time labor productivity as well as capital availability and use increased
considerably, while the overall growth rate of the economy, however, stagnated at low
levels (see Ahluwali~ 1991). Concerned about the efficiency of resource use researchers
started investigating productivity growth and input factor substitutions for aggregate
manufacturing as well as various industries. The results of these analyses differed
substantially depending on the methodology, statistical specification employed as well as
on the underlying sources of dat~ levels of aggregation and time periods considered.

Over time more sophisticated and refined methodologies in connection with longer time
series were employed to study productivity change. The contribution of total factor
productivity to output growth was of primary interest to explain the still low economic
development. Partial factor productivity was investigated to better understand the
importance of each factor of production and to evaluate substitution possibilities. In this , I
context the role of energy within the production process received increasing attention and
consequently besides the primary factors of production (capital and labor), energy and
materials were added as secondary input factors into the analyses.

Commonly, three major growth accounting approaches are considered for estimating total
factor productivity as well as total productivity growth: the Translog Index, the Solow
Index and the Kendrick Index. Total factor productivity growth (TFPG) measures the
growth in gross value added (GVA) in excess of the growth of a weighted combination of
the two inputs capital and labor. For measuring output in form of gross value added all
intermediate inputs are deducted. Thus, gross value added only provides the value that is
actually added in the production process by using the two primary inputs of production:
capital and labor. Total Productivity Growth, in contrast, relates gross value of output
(VO) to the four input factors capital, labor, energy and materials. Since it accounts for
intermediate inputs as well as primary inputs, value of output provides the more
appropriate output measure if interested in analyzing energy and material as well as
capital and labor.

The three indices developed differ in their complexity and the underlying economic
assumptions. A detailed derivation of the three indices is provided in a survey report by
Mongia and Sathaye (1998a). The Kendrick index is easy to understand in using an
arithmetic aggregation scheme for the inputs. It is restrictive in that it is based on the
assumption of a linear production function and in assigning constant (base year) shares in
GVA (VO respectively) to the inputs. The Solow index is slightly more general in
assuming a neo-classical, Cobb-Douglas, specification of the production function with
constant returns to scale, perfect competition in the market and factors being rewarded

16
 their marginal products. The translog measure is based on a more complex production
function associated with only a minimum numbers of assumptions. It is therefore of more
general nature and provides the preferably used measure for productivity growth.

Partial factor productivity (PP) indices are reported for all input factors. They are
obtained by simply dividing the value figure for each factor by the gross value of output
or by the gross value added respectively. Partial factor productivity growth indicates how
much output changes in relation to a fixed amount of each single input. It measures how
“productive” a factor is. Taking the inverse it means how much of a factor has to be used
to produce a specific amount of output - it measures the factor intensity of production.
Changes over time indicate a shift in production towards more intensive use of one factor
probably accompanied by less use of another factor. Additionally, the capital labor ratio
(K-L ratio) shows how much capital per head is used in the production process and
provides a rough measure of the capital intensity of production. The tradeoff between
capital and labor is particularly interesting in the context of labor intensive developing
countries, like Indizq that decided on the emphasis of capital intensive industries in its
early development stages in order to improve the overall economic situation.

Considering capital and labor productivity one should keep in mind that conceptually, in
situations where capital intensity is increasing over time, the analysis of partial
productivity changes may overstate the increase in labor productivity and understate the
increase in capital productivity (Ahluwali% 1991). With rising capital labor ratio
resources may shift from labor to the use of capital. Due to this shift, the measured
increase in labor productivity may be larger than the pure increase in the productivity
component (i.e. the change that is solely due to learning, learning-by-doing, improvement
of skills, experience etc.). Similarly, the increase in pure capital productivity may be
higher than the measured increase.

The next section will give an overview of previous studies that have been conducted on
productivity changes in the fertilizer industry. Thereafter, in the following section, we
develop our own estimates for both total and partial productivity using a consistent
theoretical and empirical framework.

3.1.1 Previous Studies

Previous results for statistical estimates of total factor productivity using the Translog,
Solow and/or Kendrick index as well as measures of partial factor productivity and
production fhnctions for the fertilizer industry are given in Appendix A. Figures 3.1 -3.4
dispIay both the historical as well as our own estimates graphically. The graphical
presentation allows to immediately realize the large differences in the estimates obtained
by researchers for various points of time. The overview draws on Mongia and Sathaye
(1998a).

17

-.-.--–m .. .’.. ,. .... .. ..’....,.,

,,.,,,... ,, .,.,!..~.,..,7.. ..... . ...................’.. .,’,,,. ..> .,,A,.
J-. . ..,,., ----
3.1.1.1 Partial Productivity

Capital Productivity

Partial productivity growth estimates for capital are presented in Figure 3.2. As can be
seen fi-om the figure, estimates differ depending on the time period and range under
consideration. All previous estimates reveal negative capital productivity change.
Productivity loss for Arora is lowest at –0.6Y0 p.s. for the period 1973-81. Ahluwalia
shows a slightly higher decline at –1 0/0p.s. for the period 1960-85. Gupta conducted a
study relying on primary plant level data of all chemical fertilizer plants commissioned
before 1975 distinguishing public and private plants. For the period 1969-76, Gupta
reveals a much higher decrease of capital productivity at –9.3°/0 p.s. for the private sector
and –5.6°/0 p.s. for the public sector. Bansal’s results are similar. Between 1986-94 capital
productivity declined at –3. 1% p.s. in the public sector and at –13.4% p.s. in the private
sector.

Labor Productivity

Historical estimates for labor productivity as shown in Figure 3.3 range widely from
positive to negative numbers. Ahluwalia estimates a significant increase in labor
productivity at 6.3% p.s. for the years 1960-85. Arora reports slightly lower productivity
increase at 5.4°/0 p.s. (period 1973-81). Gupta and Bansal, however, report negative
numbers for both the public and private sector. Labor productivity declines at –3 .3°/0p.s.,
–2.2% p.s. respectively, for the public sector. An even stronger decline is observed in the
private sector at -6.2% p.s., -4.8% respectively.

Capital-Labor Ratio

The overall trend in the fertilizer industry has been towards capital deepening as indicated / I
by the development of the capital-labor ratio. All studies support this finding. Both,
Ahluwalia and Bansal (private sector) conclude a considerable increase in the capital
labor ratio of 7.5% p.s. between 1960-85 and 8.6% p.s. between 1986-94 respectively.
Arora obtains a more moderate increase of 5.5% between 1973-81, while all other
estimates are in the lower range of 0.9°/0 p.s. to 3.2°/0p.s.

Material Productivity

Only few authors consider additional inputs and productivity changes in their
investigations. Gupta includes material inputs in the productivity analysis and concludes
very reverse results for the public and the private sector. A positive growth in material
productivity of 3.9% p.s. can be found in the public sector between 1969-76. However, at
the same time material productivity in the private sector fell by –2.O’XO
p.s..

18
Figure 3.1: Estimates of Total Factor Productivity Growth

Growth
(% p.s.)
Own Estimate (Kendrick)
4.4

3.4

2.4 -1 . . . . k
Own E’sthsate (Solow)
1.4
~slog)
0.4

-0.6
~.rans109)
-1.6

-2.6
Gupta (Kendrick, public)
-3.6

-4.6

-5.6

-6.6

-7.6
“

-6.6
Gupta (Kandrick, prtiate)
-9.6

-10.6
Banaal (Translog, privata)
-11.6 ,. 4=J$=A Year

Figure 3.2: Estimates of Partial Productivity Growth: Capital

Own Estbmalea (VO)

Own Estiiatas (GVA)

,- %
Bansal (public)

Year

.. . ..
Figure 3.3: Estimates of Partial Productivity Growth: Labor

Growth
(% p.s.)

7.5 ~ Own Esbm ales (VO)

6.5 . Ahluwalia

Own Estim ales (GVA)

5.5- g..~..~~~+~;z;,.,.
‘AFira
4.5 L

Year

-1.5 L
Bansal (prwale)
e-a .s-.?.
-2.5 _
Gupta (public)
..-... --. ...4.......+-
-3.5 +
1
-45j Bansal (public)

-5.5 -
Gupta (private)
t --- - --- -- -- ----
-6.5 1

Figure 3.4: Estimates of Capital-Labor Ratio

Growth
(% p.s.)

9.
.5

8.
Ahluwalia
,
I
7,
T
I

5.

Gupta (private)
. ..>
3 .!
Own Estimates
1 —-.—- .- ti -5 <
2: Gupta (public)

Bansal (pubhc)
-—

.: ... ; d—+..—+ _
,.. —.
Year

Note:“Own Estimates” are compound growth rates forthetimeperiodunderconsideration.

For the translog indices they present
exponential growth.
3.1.1.2 Total Factor Productivity Growth

Total factor productivity in the fertilizer sector has been investigated in the same studies.
Ahluwalia reports the only positive growth at an average 1.3% p.s. (translog index) for
the period 1960-85. Arora as well as Gupta and Bansal reveal losses in productivity.
While these are relatively low for Arora and the public sector analysis by Gupta and
Bansal - ranging from -0.7Y0 p.s. to –2.9Y0 p.s. – the private sector analysis reveals a
considerable decline at –8.5°/0 p.s. for the earlier period, 1969-76 (Gupta), and –1 1.5°/0
p.s. for the later period, 1986-94 (Bansal).

3.1.2 Own Estimates

In this section we present in detail our own estimates for both total and partial
productivity. We develop the Translog, Solow and Kendrick index using a consistent
theoretical and empirical fiarnework. With the recognition of energy as a critical factor
for economic growth and the special emphasis on energy use within this report, we
explicitly account for energy in using a four factor input approach (lC,L,E,M). As a
comparison, we additionally state the results obtained from the two input factor model.
Data has been compiled for the years 1973-93 from the Annual Survey of Industries,
Government of India (various years). The methodology is explained in detail in Mongia
and Sathaye (1 998).

3.1.2.1 Partial Productivity

Table 3.1 gives partial productivity based on both value of output and gross value added.
The table indicates the growth rate over the whole time period as well as split up by
different time ranges within this period. Growth rates for the time periods are calculated
as compound growth rates and time trends. This is to be in accordance with existing
growth estimates conducted by various authors and presented in section 3.1.1. above.
Figure 3.5 displays the partial productivity of capital, labor, energy and material in
relation to the value of output.

Table 3.1: Partial Productivity Growth (selected time periods, per cent p.s.)
.. .. . ..
., ”,,>,,.
,. ‘. Capitil’ ‘;,’; Labor Energy” ‘: .~~tti~al K.LL ratio- Capital Labor
.,yo/K
Gro’$h-~ ; ~ ‘“; “: ..vo./L., .V,0J13 . .:SO/~ “. , .kj’L ‘-.-:. GtiKLk GVA IL
1973-93 4.80 7.32 1.86 0.77 2.40 3.68 6.17 I
1973-79 4.68 8.45 2.00 3.43 3.61 -1.04 2.52
1979-91 8.53 10.94 2.89 0.36 2.22 7.70 10.09
1991-93 -14.73 -14.79 -4.52 -4.55 -0.07 -5.06 -5.13
Trend Rate
1973-93 7.21 8.97 2.57 1.89 1.76 6.19 7.94
Note CompoundGrowth;Trend Rate calculatedas semi-logarithmictime trend, significanton 5% level.

Over the whole time period (1973-93) all factor productivities show an increasing trend.
The growth rates as well as the figure support changes in average productivity in the late
1970s and again in the early 1990s. Generally, labor and capital follow very similar

21

. .. . . . .. . . .. . . . . .. . . . .. .. . . —.-. —
patterns as do energy and material. Between 1973 and 1979, all input factor productivities
show progress although for the years 1973-75/76 a downward trend can be observed.
From 1979 until 1990/91, labor and capital productivity advance with slight fluctuations
at high rates. Labor productivity gains are highest at an average 10.94°/0 p.s. during that
period followed by 8.53’%0p.s. growth in capital productivity. A small local peak in both
capital and labor productivity can be obsemed in 1984. In contrast, material and energy
productivity do not show such a clear upward move for the period 1979-1990/91. Until
1986/87, energy as well as material productivity remains quite stable only increasing
thereafter. Growth in material productivity is lowest at 0.36% p.s. between 1979-91,
while energy productivity increases at an average 2.89°/0 p.s. A sharp turnaround in
productivity development takes place in 1990/91. The period beginning 1990/91 exerts a
significant cut in productivity for all input factors, particularly for capital and labor (-
14.73’XOp.s. and -14.79Y0 p.s. respectively).

Figure 3.5: Index of Partial Productivity (KLEM and Value of Output)

based on 1973-74 constant values
Index

~4 I —— Year
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1965 1986 1987 1988 1989 1990 1991 1992 1993

The examination of capital and labor in relation to gross value added rather than gross
value of output confirms the results for capital and labor productivity. The results are
similar in terms of direction and size of change indicating an overall positive
development of partial productivity that was mainly supported by enormous increases in
productivity in the 1980s. The increase in labor productivity is to some extent the result
of the process of capital deepening, the increasing use of capital per head, indicated by a
growth in the capital labor ratio at 1.76% p.s. Resources have shifted from labor to the
use of capital over time.

22
3.1.2.2 Total Factor Productivity

Total factor productivity relates the input factors capital and labor to gross value added. It
measures the growth in gross value added (GVA) that can not be explained by the growth
of a weighted combination of the two inputs capital and labor.

Figure 3.6 shows the development of total factor productivity as measured by the
Kendrick, Solow and Translog Index over time. In addition, Table 3.2 gives total factor
productivity growth for different time periods. The growth rates for the Kendrick and the
Solow index are estimated as compound growth rates. The Translog index, however, is
based on the assumption of exponential growth due to its logarithmic, non-linear nature.
Trend rates calculated as semi-logarithmic trends are also given.

Table 3.2: Total Factor Productivity Growth

(selected time periods, per cent p.s.)
Growth . Translog. .SOIOW ,-;,:.: Kendtick>%,j.
1973-93 4.19 2.44 4.24
1973-79 -0.10 -1.48 -0.16
1979-91 7.92 6.19 8.19
1991-93 -5.35 -7.16 -5.07
Time Trend
1973-93 6.64 4.71 6.59
Note Tmnslog ExponentialGrowth;Solow,Kendrick CompoundGrowth.
Trend Rate calculatedas semi-logarithmictime trend, significanton 5~olevel.

Figure 3.6: Index of Total Factor Productivity

based on 1973-74constant values
Index

3.0 ~

2.6- -

2.2

1.8

i +Translog
1
1.4

i
1.0

—
0.6 1 Year
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

23
The three indices are related in their patterns, particularly the Translog and Kendrick
index that are almost identical in value and growth. The Solow index is continuously
lower both in absolute and relative terms. Similar to the evolution in partial productivity
one can observe an overall positive development of total factor productivity between
1973 and 1993 indicated by an average growth of 6.64% p.s. (Translog), 4.71% p.s.
(Solow) and 6.59% p.s. (Kendrick). The progress is mainly driven by the period of
success in the 1980s. As with labor and capital productivity total factor productivity
experiences a peak in 1984, however, followed by a dotidl in 1985 and 1986.

The first period, 1973-79, shows an overall slightly negative growth due to decreases in
total factor productivity in the first three years as well as around 1979. However, with the
beginning of the 1980s this trend reverses and total factor productivity advances at fast
rates (Translog: 7.92°/0, Kendrick: 8. 19°/0 and Solow: 6.190/0). Following policy change
towards decontrol of the fertilizer sector a turnaround can be observed in 1992. Between
1991 and 1993 all three indices reveal falling factor productivities at rates ranging from -
5.07% p.s. (Kendrick) to -7.16% p.s. (Solow).

3.1.2.3 Total Productivity

Total productivity measures the growth in gross value of output in excess of the growth
of a weighted combination of the inputs capital, labor, energy and material. As with total
factor productivity we consider three different indices for measuring total productivity.

Table 3.3 and Figure 3.7 present the growth of the three indices and their evolution over
time. Total productivity shows a more stable pattern than total factor productivity. Over
the total period of 1973 to 1993, all three indices are increasing at very similar rates. The
Translog index gives highest growth at an average 3.81 YOp.s., while the Kendrick index
is lowest at an average 3 .36°/0 p.s.

Table 3.3: Total Productivity Growth

(selected time periods, per cent p.s.)
Growth Translog Solow Kendnck
1973-93 2.31 2.16 2.05
1973-79 3.87 4.10 3.95
1979-91 3.32 3.04 2.59
1991-93 -8.44 -8.30 -6.42
Time Trend
1973-93 3.81 3.68 3.36
Note: Translog: Exponential Growth; Solow, Kendrick: Compound Growth.
Trend Rate calculated as semi-logarithmic trend, significant on s~o level.

Besides a decrease in total productivity in the first three years, the first subperiod reveals
a strong increase in total productivity. Total productivity growth ranges from 3. 87°/0 p.s.
(Translog) to 4.10% p.s. (Solow) for the period 1973-79. The second subperiod, the
period of success (179-91), experiences a minimally smaller growth at an average 2.59%
p.s. (Kendrick) to an average 3.32% p.s. (Translog). As with total factor productivity

24
growth a peak in total productivity can be observed in 1984 followed by a decline in the
two subsequent years.

Total productivity culminates in 1990 when the trend abruptly reverts to a very negative
development. The period of 1991-93 shows a sharp decrease at -6.42% p.s. for the
Kendrick index, -8.30% p.s. for the Solow index, and –8.44% p.s. for the Translog index.

Figure 3.7: Index of Total Productivity

based on 1973-74 constant values
Index
2.1

1.9

1.7

1.5

1.3
~

1.1
i

0.9

, Year
0.7 j
1973 1974 1975 1976 1977 1976 1979 1980 1981 1982 1963 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

Decomposition of Growth in Value of Output

A very insightful way of looking at growth in output is to decompose growth into the
contribution of factor input changes and total productivity growth. Generally, growth in
production is two-folded consisting of increased use of inputs and some additional
change (gain or loss) in productivity. As mentioned, growth in productivity thereby
includes technological change, learning, education, organization and management
improvements etc. The two-folded base of growth in output can naturally imply that
growth in output is accompanied by increase in factor input and decrease in productivity,
by.decrease in factor input and increase in productivity or by increase in both factor input
and productivity. Table 3.4 presents the decomposition results for our study period and
the subperiods identified above.

Table 3.4 shows overall output in the fertilizer sector measured as average exponential
growth of gross output following a positive course at 10.lVO p.s. growth over the period
1973-93. The decomposition analysis reveals that productivity gain contributed 2.3 1’%

25
p.s., about one fifth, to total output growth while the remaining growth was achieved by
increases in use of factor inputs.

1973-79 16.39 0.52 3.31 7.03 1.65 12.51 3.87

1979-91 10.42 0.00 0.49 5.46 1.15 7.10 3.32
1991-93 -10.62 0.21 1.59 -3.18 -0.80 -2.18 -8.44

The first period shows a higher than average output growth (16.39% p.s.) for the years
1973-79. Total productivity increase accounts for 3.87% p.s., slightly less than one
fourth, of output growth. A bit more than 75% of output growth is due to rising input
factors (12.51% p.s.), particularly material inputs (7.03% p.s.). Notably, productivity
growth is higher than growth in either of the three other factors. The following period
shows very similar behavior. Output grows at 10.42°/0 p.s. between 1979-91. Again, the
growth can be mainly attributed to increases in total factor inputs (7.1% p.s.). The main
contributor is growth in material inputs (5.46°/0 p.s.), followed by growth in energy input
(1.15% p.s.). Productivity growth accounts for around one third of output growth
increasing at 3.32 °/0p.s. during that period.

Following the introduction of liberalization measures, output declines considerably at –

10.62% p.s. between 1991 and 1993. The fall-down is due to both losses in productivity
as well as decreases in total inputs. Again, material input is a driving factor of output
behavior. Between 1991-93, material input declines at –3. 18% p.s. However, productivity
loss presents the single largest contributor to output loss decreasing at –8.44% p.s. and
thus accounting for more than 80’% of output loss.

3.2 Econometric Analysis

3.2.1 Previous Studies

The accounting fiarnework employed for the derivation of total and total factor
productivities does not explain why factor demand changes over time. However,
understanding substitution processes between input factors and the effects of factor price
changes on input use is crucially important for determining the rate and direction of
technological change and thus productivity growth. Few researchers so far have tried to
tackle this issue in econometrically estimating production or dual cost functions and
concluding patterns and relationships between input factors.

26
 3.2.2 Own Estimates

Our results for the econometric estimation of productivity change and patterns of input
substitution are received fkom both the statistical analysis and from estimating a translog
cost fiction approach with four input factors: capital, labor, energy and material. For a
detailed presentation of the economic framework, the specifications and the resulting
estimations see Roy et al. (1998). The foIlowing tables extract from her results and
present the most important and most interesting findings to our analysis.

Our analysis focuses on productivity growth in energy intensive industries. Accordingly,

energy prices and energy price changes over time play a dominant role. Therefore, Table
3.5 presents the elasticities of the cost shares2 for each input with respect to changes only
in energy prices. The technical bias parameter is reported for all factor inputs and is
cruciaIIy important for understanding direction and rate of technological change. It
indicates which of the factors have been substantially made use of in the process of
technological change.

Table 3.5: Estimated Parameters for the Translog Cost Function Approach
,Parameter b~, “ bie,; ,,: & ‘ ‘&: ..: “b,,‘.. . ‘b;
~,Iie; “&i;..- , b: .>:-. :b~:.... .-
0.009 -0.021 -0.098 0.109 0.004 -0.003 -0.006 0.004 -0.001
t-value (0.098) (-1.135) (-2.307) (1.577) (0.989) (3,052) (-2.973) (1.492) (-0.135)
bij= elasticity of share of ith input with respect to the change in the price of jth input
bit=technical bias parameter

Regarding the cost share elasticities the table shows that the cost shares of labor and
capital decrease with rising energy prices while the cost share of material increases with
rising energy prices. However, only the value for the energy capital relationship is
significant. The parameter b~~indicates a slight but insignificant acceleration of technical
change over time. As shown in the previous section productivity in the fertdizer sector
has been increasing over time. Thus, a significant positive technical change parameter, as
expressed by a significant negative value for bt, would indicate that this advance has been
increasing over time. Changes in productivity usually aflect the input factors differently.
The technical change bias parameters here indicate a significant capital and labor savings
bias and an insignificant energy and material using bias (Table 3.6).

Table 3.6: Technical Change Bias

r., : ~~ ‘“-’Capital.” ..’ > Labor ‘ - Energy Matetial ~
f.,..,.
Technical Change saving saving using using

For the analysis of patterns of substitution and effects of price changes on the immediate
use of input factors the cross price elasticities are of particular interest. The price
elasticity shows the extent to which the input of one factor changes in response to a price
change of one other input factor. A positive cross price elasticity indicates a substitutional

2 Cost shares are defined as factor input costs over total input costs (sum of capital, labor, energy and
material costs).

27

. . . . .-c.
 relationship between the two input factors considered. It gives an increase in factor
demand of factor i due to a decrease in factor price j which itself leads to a reduction in
demand for factor j.

Table 3.7: Price Elasticities

. . . .. .... .,
. -,-,,. ., ., ,,=-lhjce’:,,,-:‘;: ~.-.:$i$J?&PJice.::.:: ‘:
.— :, : :’ ‘-i~.ce’ty< z
<-:$+’”,:;? :-. ::,,:Mce
,....,
. .. . .. ~.: “E1asti@i:.’,,’~
:.Q::S” WW@Citi:-’ : ;.”; ~“. ,. ‘&&ci@,,: ,, :~: ,,. ~. E1.mji.city.-.
w“”- -0.376 LK 0.443 EK -0.452 MK 0.190
KL 0.136 LL -0.411 EL -0.077 ML 0.016
KE -0.362 LE -0.201 EE -0.132 ME 0.167
KM 0.603 LM 0.169 EM 0.661 MM -0.374

The price elasticities are shown in Table 3.7. All own price elasticities are negative as
required by theory. The labor price elasticity is highest with –0.411, followed by capital
price elasticity, -0.376, material price elasticity, -0.374 and an energy price elasticity of
-0-132.

Table 3.8: Elasticities of Substitution - Qualitative Overview

. .,. , . .. .. . ~ Erie@’.+ .....+: - ~~La~or‘ ~~;:.. ~capital :
Material substitutes substitutes substitutes
Energy complements complements
Labor substitutes

Cross price elasticities indicate substitutional relationship for all input factors except
energy and labor and energy and capital (Table 3.8). Thus, a rise in, for example, energy
prices will lead to increased use of material inputs to substitute for the more expensive
energy input. However, capital and to a smaller extent labor will be less intensively used
as they present complementary inputs to energy. Among the input factors, the
relationship between energy and material is most elastic. A 10°/0 increase in material price
would lead to a 6.6°/0 increase in energy input while at the same time material use would
decrease by 3.7%. However, it needs to be noted that with all resulting elasticities being
relatively modest, overall input factors are only moderately elastic.

3.3 Discussion

The results described in the previous section need to be set in context of actual changes in
both structural composition and policies within the fertilizer sector over the last 20 years
to better understand the factors driving technological change and productivity growth.

As shown above, productivity in the fertilizer sector has been increasing over time.
Productivity gains were strongest in the second half of the 1980s. The split-up of the time
range into three subperiods (1973-79, 1979-91, 1991-93) is in accordance with structural
and policy changes in the sector. Two major policy changes took place in 1977-79 and
1991/92 when first the retention price system for fertilizer was implemented and later
liberalization measures including the withdrawal of all price and distribution controls for
phosphatic and potassic fertilizers were introduced.

28
Productivity gain was high (3.87% p.s.) in the first subperiod under consideration
contributing more than one third to growth in output. However, increased use of input
factors, particularly material and capital (together 10.34% p.s.), presented the main
drivers of output growth (16.39Y0 p.s.) during that time. The fertilizer industry is one of
the only energy intensive industries that did not show any significant negative long term
reaction to the two oil price shocks in 1974 and 1979, although input costs increased
exceptionally at these points of time. Short term decreases in both output and productivity
can be observed for the years 1974, 1975 and 1980 only. Policies pertaining to the
industry may have been well effective in absorbing most of the negative consequences.
These policies were directed at ensuring fertilizer availability to farmers at reasonable
prices through subsidization.

During the following subperiod 1979-91, total control and high subsidization through the
retention price system seem to have helped preserve the upward trend of the fertilizer
sector. Output grew at an average 10.42°/0 p.s. in this period, about one third due to
improvements in productivity and the remaining due to increased use of factor inputs.
Among the factor inputs increased use of material inputs contributed most to output
growth, while at the other end labor input did nearly stagnate. Most of the growth both in
output as well as productivity was achieved during the second half of the 1980s.

In the 1980s, consumption of fertilizer also stepped significantly forward, doubling from
5.52 Mt in 1980-81 to 12.55 Mt in 1990-91. Due to increased domestic production,
imports at the same time remained level at 2.76 Mt and were mostly related to the import
of potassic fertilizer that could not be produced domestically. Although the policy system
of retention prices was originally thought to be self-financing a closer look at subsidy
figures points out that subsidies possibly were the main single factor driving the immense
progress of fertilizer supply and demand. Subsidies increased more than eightiold
between 1980-81 and 1990-91 from 5,050 million Rs. to 43,890 million Rs.

The policy of retention prices included incentives to improve the economics of plants.
For example, in linking individual retention prices to a fixed norm on capacity utilization,
individual firms profits became crucially dependent on meeting the capacity utilization
rate. Moreover, assured demand by government policy encouraged firms both to filly
utilize existing capacity and to set up new- capacity. In this view, capacity utilization in
India’s fertilizer plants improved from less than 60% in 1979-80 to almost 90’%0in 1991-
92. While the development of nitrogenous plants shows a very steady path advancing
continuously over time, phosphatic fertilizer plants reveal an up and down of capacity
utilization reaching a low in 1979-80 (66Yo) and an up in 1991-92 (920A).

Further economic advances can be attributed to increased relative sizes of plants allowing
firms to appropriate economies of scale. Between 1980 and 1990, installed capacity
doubled from 5.18 Mt to 10.19 Mt while at the same time the number of plants increased
by about 40% only. Besides government interventions, these improvements are partly due

29

. . -. .. . ~.— ,- .—,,.,
—m=m- - ~- ,., , .-, . .- —.. _
.. . . . . . .. . . . . . . .. . . . . ,. . .- 1. . . ... . . . . -?- ... ,
-. .-
to the discovery of major indigenous gas reserves in the 1980s that led to the
establishment of additional high capacity gas based ammonia plants.

The last time period, 1991-93, shows a complete turnaround from the progress
experienced in the previous period. Liberalization measures such as the withdrawal of
price and distribution control for potassic, phosphatic and low analysis nitrogenous
fertilizers together with the introduction of dual pricing including a 30% price increases
of fertilizer sold to large scale farmers, and the anticipation of fhrther decreases of
subsidies showed immediate effects on production patterns. Output decreased on average
by about the same rate it had increased in the period 1979-91
(-10.62% p.s.). Due to the nature of the policy changes and subsidy schemes towards
relatively stronger support of nitrogenous fertilizers the reduction in production, import
and use of phosphatic and potassic fertilizers was significantly higher.

Total productivity declined at a high rate of –8.44% p.s. To a large extent, this decline is
due to reduced capacity utilization caused by downfall in production, rather than being a
consequence of lack of technical progress. High levels of uncertainty particularly
regarding fi.dure policies and subsidy schemes, regarding the effects on demand and
prices, and the availability of raw materials (aggravated by the Gulf War) placed
substantial burden on the industry. However, from 1994 on with resumed subsidy
payments a recovery can be observed.

Overall, the introduction of the retention price system seemed highly effective in
stimulating the fertilizer industry as supported by our analysis. However, sustaining this
growth required ever increasing amounts of government subsidies. Economic
liberalization has led to the reduction of government support in the fertilizer industry
since 1991 with immediate effects that seem rather depressing. The long term effects,
however, are far from being clear so far.

The study shows. that technical change has been biased towards relatively less use of
capital and labor inputs. A labor savings bias seems in contrast to India’s relatively cheap
and abundant labor force. However, heavy industries like the fertilizer industry are
usually not labor intensive since manual labor is substituted by machinery and other
technology. In addition, technology transfers from more industrialized countries with
high labor costs very often implicitly import a labor saving bias to India. In comparison
to other energy intensive industries in India, fertilizer production is relatively little energy
using. Moreover, the study shows that energy productivity has been improving over time.
Nevertheless, the development of energy prices is of particular interest. Our study shows
an energy using bias although at insignificant levels. An increase in energy prices through
policy or world market changes would impose relatively higher costs if the nature of the
industry’s technical progress would be significantly biased towards the use of energy.
Technological change and productivity growth would be reduced. The analysis fhrther
reveals that labor and capital inputs are complementary to energy use. An increase in
energy prices would therefore reduce demand for labor and capital. However, inter-input
substitution possibilities are limited.

30
4. Future Development of the Fertilizer Sector

4.1 Ongoing Changes in the Fertilizer Industry

The fertilizer industry today as in the past faces major challenges. Continued supply and
use of fertilizer is important to ensure the country’s food security goal. Foodgrain
production was declining in the most recent past (e.g. between 1994-96) potentially
jeopardizing this goal (Government of Indi% Department of Agriculture and Co-
operation, 1999; Trivedi, 1998). The industry’s challenges relate to the uncertainty in the
supply and pricing of feedstock, especially of natural gas and naphtha, low efficiency and
small size of older plants, high investment costs, in?hstructural bottlenecks and an
uncertain policy environment.

The current composition of ammonia plants in terms of age, capacity and feedstock is
presented in Table 4.1. The table shows that plants installed before 1980 are small in size
using less efficient feedstock. Plants installed in the 1980s are relatively large and employ
more efficient feedstock whereas plants built during the 1990s are nearly world-scale in
terms of feedstock use.

Table 4.1: Age, Number, Installed Capacity and Feedstock of Ammonia Plants in
India
“Year,ofIi&dlation . .. .1,-: 1951 to,1970 } :. 1971to 1980 “.;.:J981to 3$?90? ;: ‘:19910
nwards
Number of Plants 10 15 11 4
Share in Total 25% 38% 28% 1o%
Aggr. Installed Capacity (tpd) 4622 10815 14930 3872
Share in Total 1470 32% 44% 11%
Average Plant Size (tpd/plant) 462 721 1357 968
Feedstock Demand (tpd)
NGIAG 1000 1510 10950 3872
Naphtha 2722 5420 1730
Fuel Oil/LSHS 900 2885 1350
Coal 900 900
Source: Trivedi (1998).

Natural gas is the preferred feedstock to the fertilizer industry. Gas based plants are more
efficient than plants based on other feedstock and thus produce at lowest cost. Although
under government control feedstocks were individually assigned to plants, gas based
fertilizer plants have been supplied with gas barely sufilcient to meet the feedstock
requirements. In addition, the calorific value of gas has been decreasing. Demand of gas
from the fertilizer industry is in competition with demand from the high priority power
sector. In the process of industrial liberalization, the government intends to eventually
decontrol the prices
. of various feedstocks. Price and quantity concessions to the fertilizer
industry might then be withdrawn leading to price increases for supply to the fertilizer
industry much higher than to other industries in order to reduce the existing price

31

——
—-.
differential. With respect to the food security goal this would lead to a need for
tremendously increased subsidies.

In view of the uncertainty regarding the availability of feedstock and due to recent policy
changes towards liberalization and reduction of subsidies, no investment into setting up
new capacity has been taken since the early 1990s. Additions in capacity in the 1990s
resulted from projects planned in the 1980s but commissioned in the 1990s. However,
revamping and modernization options are being explored. In order to reduce production
costs capacity utilization and energy consumption need to be improved. Energy saving
technological developments are being commercialized and technologically advanced
processes are being adopted. Furthermore, joint venture projects are being envisioned, in
particular for phosphatic fertilizer where production is completely based on imported raw
materials.

The government has constituted a new fertilizer pricing policy review committee (HPC)
that was asked to review the existing system of subsidization of urea and suggest an
alternative broad based scientific and transparent methodology and recommend measures
for greater cohesiveness in policies. Policies have led to a currently heavily imbalance
use of fertilizers through decontrolled high price potassium and phosphatic fertilizers and
artificially low urea prices. Furthermore, urea imports reached a peak of 3.77 million
tonnes in 1995-96 contributing further to the imbalance in fertilizer use. (National
Information Center, 1999)

4.2 Potentials for Energy Efficiency Improvements

4.2.1 India versus Best Practice

Table 4.2 presents energy savings potentials by comparing specific energy consumption
in Indian ammonia plants with specific energy consumption that could be achieved if
particular revamp and modernization efforts would be undertaken. A typical energy
efficiency revamp of a plant would reduce specific energy consumption for plants
installed before 1980 by 1.2 to 3.2 Gcal/t and for plants installed between 1981-1990 by
about 0.8 to 1 Gcal/t depending on the feedstock used. Plants installed after 1991 are
already highly efficient and meet world best standards. Currently, there are only two coal
based plants accounting for a small share of installed capacity. They are high energy
consumers and do not promise significant energy saving potentials. With a production
capacity of 11.3 Mt per year, overall energy savings amount to 14.12 Peal per year
(Trivedi, 1998).

The age of the technology, the scale of the plant and management practices have a large
impact on energy efficiency of the overall process. Energy savings potential are highest
for naphtha and fuel oil/LSHS based plants especially for plants built before 1980. A
closer look at all pre-1980 plants reveals that with current installed capacity of 5.09 Mt,
energy savings in the pre-1980 ammonia plants alone would account for up to 11 Peal per

32
 year. This means, with a share of less than 50V0 in total installed capacity energy savings
in these plants alone would be more than 75°/0 of total possible energy savings.

Table 4.2: Energy Savings Potential -All Ammonia Plants: Broad Estimates
Year of Installation----: -’=i-:~.~:~.., ‘{Na~ralGas~., ‘:, Naph@ ~~-;J;.’3?~el,’0ill””j,, “’:~”~’ “’Coa]’t. .’
,; ..’-. ---- ,.
....,... :.. .,.. . As:ocjated Gas ‘:.: ‘-? .; :..Z:.!$”:.“-”ys~s:. ,; “-’’.~” ‘-:-.-.-;;-..”;,
-.”’
Total Installed Capaci~ (tpd) 17332 9872 5235 1800
Production Capaci~ (A4t/year~, 5.72 3.26 1.73 0.59
installed in
Pre-1980 Plants 0.83 2.69 1.28 0.30
Post-1980 Plants 4.89 0.57 0.45 0.29
Specific Energy Consumption
(1993-94) (Gcal/t ammonia)
Pre-1980 10 12 15 38
1981 – 1990 9.7 10.6 11.8 48
1991 – today 8.7
Average 9.36 11.53 14.17 39.70
Achievable Specific Energy
Consumption (Gcal/t ammonia)
Average 8.8 9.8 11.8 38.0
Energy Savings Potential (%)
Average 6% 1570 17% 4%
Total Average Energy Savings 3.43 5.54 4.15 1.00
(Peal/year)
Source: Trivedi (1998).
. .
The maximumoperationperiod of a plant is 330 days a year. baveragefor all ammoniaplants.

The scale of the plant matters significantly for energy consumption. Therefore revamp
options (as described in more detail below) typically include a capacity increase for the
plant. This will help meeting fkture demand which is expected to grow at 3V0 annually
between 1997 and 2001.

4.2.2 Categories for Energy Efilciency Improvement

The following factors have already been identified as aflecting energ consumption in
ammonia-urea plants: capacity utilization, type of feedstock, technology employed, and
vintage of the plant. Since ammonia-urea plants built more recently already reach world-
best efilciency levels focus here is placed on pre-1980s and 1980s plants. For an older
ammonia plant a typical revamp would include the following: a) capacity increase, b)
energy-saving, c) reduction in raw material and utility consumption, d) reduction in
environmental impact, e) improved safety and reliability, and f) improved control
systems. All of this would directly or indirectly benefit energy consumption.

Specifically, improvements to the energy efficiency of various process components of

ammonia (and to a lower extent of urea) include improvements in the reforming, COZ
removal, synthesis, and purge gas recovery (all of these can fimther broken down into
process optimization, maximization of heat recovery, and the fine-tuning of process
parameters), and energy savings from development of new ~d better catalysts ~d

33

....- - ..—-. -r ..-,. ,?., , . . . . ..x — . ..mr..a.r, ., —.- . .. . —., ~=

-., . . . .
improved metallurgy leading to superior and more efficient process equipment. For a
more detailed description of these measures and their actual energy savings potentials see
Trivedi (1998).

4.2.3 Barriers to Energy Efllciency Improvement

Although integrating energy savings measures and technologies in all plants would lead
to net savings both in terms of energy and overall costs, only few measures have been or
are currently being implemented in the Indian fertilizer sector. Barriers to energy
efilciency improvement are of both general and finrdprocess specific nature. A 5-6 year
payback period is one of the barriers to energy-saving investment.

At the macro level, government policy towards liberalization and the pressure to reduce
subsidies create uncertainty and pose challenges to adoption of more energy efficiency.
As energy prices were regulated in the past and feedstocks assigned to the plants, energy
costs were almost taken as fixed by the plant and little incentive was prevalent to reduce
energy consumption. The energy price elasticity as shown in section 3.2.2 was low at –
0.13. However, one should keep in mind that in a regulated energy market for fertilizer
production, price elasticities do not filly display a firm’s actual behavior to price signals
as they would do in a deregulated, truly competitive market. Therefore, energy price
signals may be more effective in stimulating energy conservation than concluded from
the economic analysis. Further, firm-level barriers to the adoption of energy efficiency
improvement include significant capital requirements, lack of appreciation for reducing
energy use and familiarity with the technical and commercial aspects of energy efficiency
and environmental management.

Although in view of the uncertainties regarding the policy environment and feedstock
supply, revamping programs present the preferred option over setting up new plants, there
are still many barriers related to these. Limitations imposed by existing plant layout and
technology and utility systems in use may rule out the possibility of introducing new
developments. Moreover, pre-revamp activities such as identi~ing bottlenecks,
evaluating available technological options and their commercial impact, assessing plant
layout and constraints of the existing set-up are very complex and require a well
coordinated effort amongst various disciplines. (Trivedi, 1998)

5. Summary and Conclusions

In this paper, we investigated India’s fertilizer sector from various perspectives. We

developed economic as well as engineering indicators for productivity growth, technical
change and energy consumption that allowed us to investigate savings potentials in
specific energy use. We discussed our findings within a broader context of structural and
policy changes in the sector. The economic analysis showed that productivity was
increasing over time. The increase took place during the era of total control when the
retention price system and distribution control was in effect. The retention price system
was coupled with relatively high norms on capacity utilization which supported

34
productivity increase. With liberalization of the fertilizer sector and reduction of subsidies
productivity declined substantially since the early 1990s, despite an increase in capacity
utilization.

We fbrther pointed out options for reducing energy consumption in the fertilizer industry.
In comparing current energy consumption to best achievable energy consumption energy
savings of up to 17°/0 could be achieved. However, the implementation of initiatives
towards energy efficiency is being hampered by bmriers both of general and process
specific nature occurring at the macro and micro level of the economy.

Energy policies in general and price-based policies in particular can help overcome these
barriers in giving proper incentives and correcting distorted prices. Appropriate
provisions should be made in the retention pricing scheme to tier encourage
investment in energy conservation projects. Originally, normative consumption of various
inputs was taken into account under the retention price system which encouraged the
implementation of energy efficiency measures. These and other fiscal incentives need to
be reinstated under the current scheme.

Through the removal of subsidies energy prices would come to reflect their true costs. In
a deregulated market, firms would adjust their behavior in order to minimize costs of
production. Therefore, in the short term, energy price increases would push less
productive and ineftlcient mostly smaller units out of the market or force these units to
take immediate initiatives to improve productivity and efficiency. On a long term basis,
substantial fhrther investments in energy efficiency technologies for existing and new
plants have to be made. Therefore, sectoral policies should be devoted to the promotion
of such investments. A stable foreseeable policy environment would substantially help
firms to reduce the risk of taking large investments. Our economic results pointed out that
technological change has been biased3 towards the use of energy inputs. This implies that
price-based policies albeit effective in reducing energy use could have a negative long run
effect on productivity, and thus welfare. An optimal policy strategy would therefore
consist of a mix of regulatory and price based incentives within a set political and
economic framework.

3significant at 15’%level.

35
References

Ahluwali~ Isher Judge, 1991: Productivity and Growth in Indian Manufacturing, Delhi,
Oxford New York: Oxford University Press.

Ahluwali~ Isher Judge, 1985: Industrial Growth in India – Stagnation Since the Mid-
Sixties, Delhi, Oxford, New York: Oxford University Press.

Basu, P. and Madhukar H. Majmudar, 1994: “Policy Support for the Phosphatic Fertiliser
Industry in India”, in Challenges of Liberalisation in the Fertiliser and Agriculture
Sectors, The Fertilizer Association of India, FAI Seminar December 1994, New
Delhi, India.

Center for Monitoring the Indian Economy (CMIE), 1996: India’s Industrial Sector,
Economic Intelligence Service, India, January, 1996.

Datt, Ruddar and K.P.M. Sundharam, 1998: Indian Economy, New Delhi, Chand &
Company Ltd.

Government of India, 1999: Fertilizer Situation in lMDIA, Department of Agriculture and

Co-operation, httu ://www.nic.k/agri, National Informatics Centre (NC).

Government of India (various years): Annual Survey of Industries: Summary Results for
Factory Sector, 1973-93, Central Statistical Organisation, Department of Statistics,
Ministry of Planning and Programme Implementation, New Delhi, India.

Kalra, G.D., 1989: Electiveness of Incentives for Energy Saving Devices, New Delhi,
India: National Council of Applied Economic Research.

Mittal, D.K., 1994: Fertiliser Indust~, New Delhi, India, Anmol Publications Ltd.

Mongia, Puran B. and Jayant Sathaye, 1998: Productivity Trends in India’s Energy
Intensive Industries – A Growth Accounting Analysis, Lawrence Berkeley National
Laboratory, 41838, Berkeley, California.

Mongia, Puran B. and Jayant Sathaye, 1998a: Productivity Growth and Technical
Change in India’s Energy Intensive Industries – A Survey, Lawrence Berkeley
National Laboratory, 41840, Berkeley, California.

National Information Center, 1999: Fertilizer Situation in INDIA, Department of

Agriculture & Cooperation, Government of India, h~://www.nic.in/a~ri/fertsit/htm.

Phylipsen, G.J.M., K. Blok, and E. Worrell, 1998: Handbook on International

Comparisons of Energy E@ciency in the Manufacturing Industry, Dept. of Science,
Technology and Society, Utrecht University, The Netherlands.

36
Prasad, Rajendra and J.F. Power, 1994: “Balanced Fertilisation and Sustainable
Agriculture in the Wake of Recent Policy Changes”, in Challenges of Liberalisation
in the Fertiliser and Agriculture Sectors, The Fertilizer Association of Indi~ FAI
Seminar, December 1994, New Delhi, India.

Roy, Joyashree, J. Sathaye, A. Sanstad, P. Mongia, and K. Schumacher, 1999:

“Productivity Trends in India’s Energy Intensive Industries”, The Energy Journal,
Vol. 20, No. 3, July.

Singh, Duleep and A.P. Srivastava, 1994: “Approach and Outlook of Likely Policy
Environment of Nitrogenous Fertiliser Industry in India”, in Challenges of
Liberalisation in the Fertiliser and Agriculture Sectors, The Fertilizer Association of
India, FAI Seminar, December 1994, New Delhi, India.

Subbiah, K. S., 1994: “Need for Effective Demand/Supply Management under

Decontrolled/Decanalised Scenario”, in Challenges of Liberalisation in the Fertiliser
and Agriculture Sectors, The Fertilizer Association of Indi% FAI Seminar, December
1994, New Delhi, India.

Tandon, H.L.S. and Pratap Narayan, 1990: Fertilizers in Indian Agriculture: Past,
Present and Future ~1950-2000), Fertiliser Development and Consultation
Organisation, New Delhi, India.

TERI, 1994: Teri Energy Data Directo~ and Yearbook 1994/95, Tata Energy Research
Institute, New Delhi, India: Pauls Press.

TERI, 1996: Teri Energy Data Directory and Yearbook 1996/97, Tata Energy Research
Institute, New Delhi, India: Pauls Press.

Trivedi, A.N., J. Sathaye, A. Gadgil, M. Mukhopadhyay, 1998: Energy Eficiency and

Environmental Management Options in the Indian Fertilizer Industry, ADB
Technical Assistance Project (TA:2403-IND), Forest Knolls, Calif.: ERI.

World Energy Council, 1995: “Energy Efficiency Improvement Utilising High

Technology: An Assessment of Energy Use in Industry and Buildings”, prepared by:
Marc D. Levine, Lynn Price, Nathan Martin and Ernst Worrell, London: World
Energy Council.

Worrell, E., R.J.J. van Heijningen, J.F. de Castro, J.H.O. Hazewinkel, J.G. de Beer,
A.P.C. Faaij, and K. Vringer, 1994: “New Gross Energy-Requirement Figures for
Materials Production”, Energv, Vol. 19 No. 6.
Appendix

Appendix A

Fertilizer
- -. . . . . . . Historical
----.----—. l?stimates
—--- ——.-.
A.utior
. . . .. “’:----“ “.:‘:
‘ Nfeti@!?@@fsj~::’. .Wiiice”of: “ ‘PefioA@: GroWhRate’
,..:.”. : ., :...,, ,.,,\.~.:?~~ . ::.:::::;,
. .“. ..:..:.’,.”
. ... ..’’ .,. ; ba~ ,’-; ‘-;; ‘;:,:”’:;:;. ‘; “ ,,. .
Ahluwalia TFPG: TL ASI 1960-85 1.3
(1991) PP: Capital -1.0
PP: Labor 6.3
Cap/Lab Ratio 7.5
Arora (1987) TFPG: TL ASI 1973-81 -0.65
PP: Capital -0.62
PP: Labor 5.41
Cap/Lab Ratio 5.50
Gupta (1982) TFPG: Kendrick Plant Level 1969-76 -2.42
PP: Capital Public -5.62
PP: Labor -3.32
PP: Materials 3.85
Cap/Lab Ratio 2.30
TFPG: Kendrick- - Plant Level 1969-76 -8.53
PP: Capital Private -9.36
PP: Labor -6.20
PP: Materials -2.00
Cap/Lab Ratio 3.16
Bansal (1997) TFPG: TL Plant Level 1986-94 -2.91
PP: Capital Public -3.10
PP: Labor -2.20
Cap/Lab Ratio
. .. . . . 0.90
‘– TF-PG:TL Plant Level 1986-94 -11.53
PP: Capital Private -13.4
PP: Labor -4.80
Cap/Lab Ratio 8.60
Source: Mongia and Sathaye (1998a)
..- . . ... ., .,. ,
NOtf2: tirowm rates are percent per annum, enner compouna annual growm rates, seml+og rreno
rates or simple average growth rates.

38